We investigate some previously unexplored (or underexplored) computational aspects of total variation (TV) distance. First, we give a simple deterministic polynomial-time algorithm for checking equivalence between mixtures of product distributions, over arbitrary alphabets. This corresponds to a special case, whereby the TV distance between the two distributions is zero. Second, we prove that unless $\mathsf{NP} \subseteq \mathsf{RP}$ it is impossible to efficiently estimate the TV distance between arbitrary Ising models, even in a bounded-error randomized setting.

This paper investigates *list replicability* [Dixon et al., 2023] in the context of multi-armed (also linear) bandits (MAB). We define an algorithm $A$ for MAB to be $(\ell,\delta)$-list replicable if with probability at least $1-\delta$, $A$ has at most $\ell$ traces in independent executions even with different random bits, where a trace means sequence of arms played during an execution. For $k$-armed bandits, although the total number of traces can be $\Omega(k^T)$ for a time horizon $T$, we present several surprising upper bounds that either independent of or logarithmic of $T$: (1) a $(2^{k},\delta)$-list replicable algorithm with near-optimal regret, $\widetilde{O}({\sqrt{kT}})$, (2) a $(O(k/\delta),\delta)$-list replicable algorithm with regret $\widetilde{O}\left(\frac{k}{\delta}\sqrt{kT}\right)$, (3) a $((k+1)^{B-1}, \delta)$-list replicable algorithm with regret $\widetilde{O}(k^{\frac{3}{2}}T^{{\frac{1}{2}}+2^{-(B+1)}})$ for any integer $B>1$. On the other hand, for the sublinear regret regime, we establish a matching lower bound on the list complexity (parameter $\ell$). We prove that there is no $(k-1,\delta)$-list replicable algorithm with $o(T)$-regret. This is optimal in list complexity in the sub-linear regret regime as there is a $(k, 0)$-list replicable algorithm with $O(T^{2/3})$-regret. We further show that for linear bandits with $d$-dimensional features, there is a $\widetilde{O}(d^2T^{1/2+2^{-(B+1)}})$-regret algorithm with $((2d+1)^{B-1},\delta)$-list replicability, for $B>1$, even when the number of possible arms can be infinite.

In this work, we study the notion of representation obliviousness in the context of pseudodeterministic streaming algorithms. A (randomized) streaming algorithm A is pseudodeterministic, if for every stream D, there is a ''canonical value'' g(D) so that with probability at least 2/3, A on input stream D outputs g(D). Intuitively, a randomized algorithm is representation oblivious if the output distribution of the algorithm does not depend on the representation of the input. We investigate this notion in the context of streaming algorithms, more specifically, distinct elements estimation (F 0 estimation) in data streams. In this context, representation obliviousness captures the idea that the output distribution of an algorithm for estimating F 0 should only depend on the set of distinct elements of the stream. This is a natural notion, as we note that standard streaming algorithms are representation-oblivious in this sense. We prove that any representation oblivious pseudodeterministic streaming algorithm for estimating F 0 must use Ω(n) space, where [n] is the universe. More generally, we prove that any representation oblivious pseudodeterministic t(n)-pass streaming algorithm requires Ω(n/t(n)) space. This lower bound matches the space requirement of the straightforward multi-pass deterministic algorithm that exactly computes F 0 .

Submodular optimization has become increasingly prominent in machine learning, and fairness has drawn much attention. In this paper, we propose to study the fair k-submodular maximization problem and develop a 1/3-approximation greedy algorithm with a running time of O(knB). Our theoretical guarantee matches the best-known k-submodular maximization results without fairness constraints. In addition, we have developed a faster threshold-based algorithm that achieves a (1/3 ϵ) approximation with ${\mathcal{O}}\left({\frac{{kn}}{\varepsilon }\log \frac{B}{\varepsilon }}\right)$ evaluations of the function−f. Furthermore, for both algorithms, we provide approximation guarantees when the k-submodular function is not accessible but only can be approximately accessed. We have extensively validated our theoretical findings through empirical study and examined the practical implications of fairness. The experimental results show that the fairness constraints do not significantly undermine the quality of solutions.

Submodular optimization finds applications in machine learning and data mining. In this paper, we study the problem of maximizing functions of the form h = f-c, where f is a monotone, non-negative, weakly submodular set function and c is a modular function. We design a deterministic approximation algorithm that runs with O(n/ε log n/(γ ε) ) oracle calls to function h, and outputs a set S such that h(S) ≥ γ(1-ε)f(OPT)-c(OPT)-c(OPT)/γ(1-ε) log f(OPT)/c(OPT), where γ is the submodularity ratio of f. Existing algorithms for this problem either admit a worse approximation ratio or have quadratic runtime. We also present an approximation ratio of our algorithm for this problem with an approximate oracle of f. We validate our theoretical results through extensive empirical evaluations on real-world applications, including vertex cover and influence diffusion problems for submodular utility function f, and Bayesian A-Optimal design for weakly submodular f. Our experimental results demonstrate that our algorithms efficiently achieve high-quality solutions.

In this paper, we establish a novel connection between total variation (TV) distance estimation and probabilistic inference. In particular, we present an efficient, structure-preserving reduction from relative approximation of TV distance to probabilistic inference over directed graphical models. This reduction leads to a fully polynomial randomized approximation scheme (FPRAS) for estimating TV distances between same-structure distributions over any class of Bayes nets for which there is an efficient probabilistic inference algorithm. In particular, it leads to an FPRAS for estimating TV distances between distributions that are defined over a common Bayes net of small treewidth. Prior to this work, such approximation schemes only existed for estimating TV distances between product distributions. Our approach employs a new notion of partial couplings of high-dimensional distributions, which might be of independent interest.

This paper studies replicability in machine learning tasks from a geometric viewpoint. Recent works have revealed the role of geometric partitions and Sperner's lemma (and its variations) in designing replicable learning algorithms and in establishing impossibility results. A partition $\mathcal{P}$ of $\mathbb{R}^d$ is called a $(k,\epsilon)$-secluded partition if for every $\vec{p}\in\mathbb{R}^d$, an $\varepsilon$-radius ball (with respect to the $\ell_{\infty}$ norm) centered at $\vec{p}$ intersects at most $k$ members of $\mathcal{P}$. In relation to replicable learning, the parameter $k$ is closely related to the $\textit{list complexity}$, and the parameter $\varepsilon$ is related to the sample complexity of the replicable learner. Construction of secluded partitions with better parameters (small $k$ and large $\varepsilon$) will lead to replicable learning algorithms with small list and sample complexities. Motivated by this connection, we undertake a comprehensive study of secluded partitions and establish near-optimal relationships between $k$ and $\varepsilon$. 1. We show that for any $(k,\epsilon)$-secluded partition where each member has at most unit measure, it must be that $k \geq(1+2\varepsilon)^d$, and consequently, for the interesting regime $k\in[2^d]$ it must be that $\epsilon\leq\frac{\log_4(k)}{d}$. 2. To complement this upper bound on $\epsilon$, we show that for each $d\in\mathbb{N}$ and each viable $k\in[2^d]$, a construction of a $(k,\epsilon)$-secluded (unit cube) partition with $\epsilon\geq\frac{\log_4(k)}{d}\cdot\frac{1}{8\log_4(d+1)}$. This establishes the optimality of $\epsilon$ within a logarithmic factor. 3. Finally, we adapt our proof techniques to obtain a new ``neighborhood'' variant of the cubical KKM lemma (or cubical Sperner's lemma): For any coloring of $[0,1]^d$ in which no color is used on opposing faces, it holds for each $\epsilon\in(0,\frac12]$ that there is a point where the open $\epsilon$-radius $\ell_\infty$-ball intersects at least $(1+\frac23\epsilon)^d$ colors. While the classical Sperner/KKM lemma guarantees the existence of a point that is "adjacent" to points with $(d+1)$ distinct colors, the neighborhood version guarantees the existence of a small neighborhood with exponentially many points with distinct colors.

In today's digital age, it is becoming increasingly prevalent to retain digital footprints in the cloud indefinitely. Nonetheless, there is a valid argument that entities should have the authority to decide whether their personal data remains within a specific database or is expunged. Indeed, nations across the globe are increasingly enacting legislation to uphold the "Right To Be Forgotten" for individuals. Investigating computational challenges, including the formalization and implementation of this notion, is crucial due to its relevance in the domains of data privacy and management. This work introduces a new streaming model: the 'Right to be Forgotten Data Streaming Model' (RFDS model). The main feature of this model is that any element in the stream has the right to have its history removed from the stream. Formally, the input is a stream of updates of the form (a, Δ) where Δ ∈ {+, ⊥} and a is an element from a universe U. When the update Δ=+ occurs, the frequency of a, denoted as f a , is incremented to f a +1. When the update Δ=⊥, occurs, f a is set to 0. This feature, which represents the forget request, distinguishes the present model from existing data streaming models. This work systematically investigates computational challenges that arise while incorporating the notion of the right to be forgotten. Our initial considerations reveal that even estimating F 1 (sum of the frequencies of elements) of the stream is a non-trivial problem in this model. Based on the initial investigations, we focus on a modified model which we call α-RFDS where we limit the number of forget operations to be at most α fraction. In this modified model, we focus on estimating F 0 (number of distinct elements) and F 1 . We present algorithms and establish almost-matching lower bounds on the space complexity for these computational tasks.

Interpretations of logical formulas over semirings (other than the Boolean semiring) have applications in various areas of computer science including logic, AI, databases, and security. Such interpretations provide richer information beyond the truth or falsity of a statement. Examples of such semirings include Viterbi semiring, min-max or access control semiring, tropical semiring, and fuzzy semiring. The present work investigates the complexity of constraint optimization problems over semirings. The generic optimization problem we study is the following: Given a propositional formula phi over n variable and a semiring (K,+, . ,0,1), find the maximum value over all possible interpretations of phi over K. This can be seen as a generalization of the well-known satisfiability problem (a propositional formula is satisfiable if and only if the maximum value over all interpretations/assignments over the Boolean semiring is 1). A related problem is to find an interpretation that achieves the maximum value. In this work, we first focus on these optimization problems over the Viterbi semiring, which we call optConfVal and optConf. We first show that for general propositional formulas in negation normal form, optConfVal and optConf are in FP^NP. We then investigate optConf when the input formula phi is represented in the conjunctive normal form. For CNF formulae, we first derive an upper bound on the value of optConf as a function of the number of maximum satisfiable clauses. In particular, we show that if r is the maximum number of satisfiable clauses in a CNF formula with m clauses, then its optConf value is at most 1/4^(m-r). Building on this we establish that optConf for CNF formulae is hard for the complexity class FP^NP[log]. We also design polynomial-time approximation algorithms and establish an inapproximability for optConfVal. We establish similar complexity results for these optimization problems over other semirings including tropical, fuzzy, and access control semirings.

Total variation distance (TV distance) is a fundamental notion of distance between probability distributions. In this work, we introduce and study the problem of computing the TV distance of two product distributions over the domain {0,1}^n. In particular, we establish the following results. 1. The problem of exactly computing the TV distance of two product distributions is #P-complete. This is in stark contrast with other distance measures such as KL, Chi-square, and Hellinger which tensorize over the marginals leading to efficient algorithms. 2. There is a fully polynomial-time deterministic approximation scheme (FPTAS) for computing the TV distance of two product distributions P and Q where Q is the uniform distribution. This result is extended to the case where Q has a constant number of distinct marginals. In contrast, we show that when P and Q are Bayes net distributions the relative approximation of their TV distance is NP-hard.

We investigate replicable learning algorithms. Informally a learning algorithm is replicable if the algorithm outputs the same canonical hypothesis over multiple runs with high probability, even when different runs observe a different set of samples from the unknown data distribution. In general, such a strong notion of replicability is not achievable. Thus we consider two feasible notions of replicability called {\em list replicability} and {\em certificate replicability}. Intuitively, these notions capture the degree of (non) replicability. The goal is to design learning algorithms with optimal list and certificate complexities while minimizing the sample complexity. Our contributions are the following. 1. We first study the learning task of estimating the biases of $d$ coins, up to an additive error of $\varepsilon$, by observing samples. For this task, we design a $(d+1)$-list replicable algorithm. To complement this result, we establish that the list complexity is optimal, i.e there are no learning algorithms with a list size smaller than $d+1$ for this task. We also design learning algorithms with certificate complexity $\tilde{O}(\log d)$. The sample complexity of both these algorithms is $\tilde{O}(\frac{d^2}{\varepsilon^2})$ where $\varepsilon$ is the approximation error parameter (for a constant error probability). 2. In the PAC model, we show that any hypothesis class that is learnable with $d$-nonadaptive statistical queries can be learned via a $(d+1)$-list replicable algorithm and also via a $\tilde{O}(\log d)$-certificate replicable algorithm. The sample complexity of both these algorithms is $\tilde{O}(\frac{d^2}{\nu^2})$ where $\nu$ is the approximation error of the statistical query. We also show that for the concept class \dtep, the list complexity is exactly $d+1$ with respect to the uniform distribution. To establish our upper bound results we use rounding schemes induced by geometric partitions with certain properties. We use Sperner/KKM Lemma to establish the lower bound results.

We investigate the problems of maximizing k-submodular functions over total size constraints and over individual size constraints. k-submodularity is a generalization of submodularity beyond just picking items of a ground set, instead associating one of k types to chosen items. For sensor selection problems, for instance, this enables modeling of which type of sensor to put at a location, not simply whether to put a sensor or not. We propose and analyze threshold-greedy algorithms for both types of constraints. We prove that our proposed algorithms achieve the best known approximation ratios for both constraint types, up to a user-chosen parameter that balances computational complexity and the approximation ratio, while only using a number of function evaluations that depends linearly (up to poly-logarithmic terms) on the number of elements n, the number of types k, and the inverse of the user chosen parameter. Other algorithms that achieve the best-known deterministic approximation ratios require a number of function evaluations that depends linearly on the budget B, while our methods do not. We empirically demonstrate our algorithms’ performance in applications of sensor placement with k types and influence maximization with k topics.

We consider the problem of maximizing submodular functions under submodular constraints by formulating the problem in two ways: SCSKC and DiffC. Given two submodular functions f and g where f is monotone, the objective of SCSKC problem is to find a set S of size at most k that maximizes f(S) under the constraint that g(S) < theta, for a given value of theta. The problem of DiffC focuses on finding a set S of size at most k such that h(S) = f(S)-g(S) is maximized. It is known that these problems are highly inapproximable and do not admit any constant factor multiplicative approximation algorithms unless NP is easy. Known approximation algorithms involve data-dependent approximation factors that are not efficiently computable. We initiate a study of the design of approximation algorithms where the approximation factors are efficiently computable. For the problem of SCSKC, we prove that the greedy algorithm produces a solution whose value is at least (1-1/e)f(OPT) - A, where A is the data-dependent additive error. For the DiffC problem, we design an algorithm that uses the SCSKC greedy algorithm as a subroutine. This algorithm produces a solution whose value is at least (1-1/e)h(OPT)-B, where B is also a data-dependent additive error. A salient feature of our approach is that the additive error terms can be computed efficiently, thus enabling us to ascertain the quality of the solutions produced.

Constraint satisfaction problems (CSPs) and data stream models are two powerful abstractions to capture a wide variety of problems arising in different domains of computer science. Developments in the two communities have mostly occurred independently and with little interaction between them. In this work, we seek to investigate whether bridging the seeming communication gap between the two communities may pave the way to richer fundamental insights. To this end, we focus on two foundational problems: model counting for CSP’s and computation of zeroth frequency moments ( F 0 ) for data streams. Our investigations lead us to observe a striking similarity in the core techniques employed in the algorithmic frameworks that have evolved separately for model counting and F 0 computation. We design a recipe for translating algorithms developed for F 0 estimation to model counting, resulting in new algorithms for model counting. We also provide a recipe for transforming sampling algorithm over streams to constraint sampling algorithms. We then observe that algorithms in the context of distributed streaming can be transformed into distributed algorithms for model counting. We next turn our attention to viewing streaming from the lens of counting and show that framing F 0 estimation as a special case of #DNF counting allows us to obtain a general recipe for a rich class of streaming problems, which had been subjected to case-specific analysis in prior works. In particular, our view yields an algorithm for multidimensional range efficient F 0 estimation with a simpler analysis.

A probabilistic algorithm A is pseudodeterministic if, on every input, there exists a canonical value that is output with high probability. If the algorithm outputs one of k canonical values with high probability, then it is called a k-pseudodeterministic algorithm. In the study of pseudodeterminism, the Acceptance Probability Estimation Problem (APEP), which is to additively approximate the acceptance probability of a Boolean circuit, is emerging as a central computational problem. This problem admits a 2-pseudodeterministic algorithm. Recently, it was shown that a pseudodeterministic algorithm for this problem would imply that any multi-valued function that admits a k-pseudodeterministic algorithm for a constant k (including approximation algorithms) also admits a pseudodeterministic algorithm (Dixon, Pavan, Vinodchandran; ITCS 2021).

We investigate the problem of multi-objective submodular optimization with cardinality constraint in the context of δ-approximate oracle and show that it is possible to ensure (1 − 1/e)2− 3δ-approximate guarantee for the multi-objective submodular optimization problem. We show that group influence maximization in online social networks is an instance of this optimization problem with cardinality constraint and δ-oracle. We develop a prototype implementation of our solution strategy for group influence maximization problem for networks of different sizes and experimentally justify the effectiveness and scalability of our strategy.

We exhibit several computational problems that are complete for multi-pseudodeterministic computations in the following sense: (1) these problems admit 2-pseudodeterministic algorithms (2) if there exists a pseudodeterministic algorithm for any of these problems, then any multi-valued function that admits a k-pseudodeterministic algorithm for a constant k, also admits a pseudodeterministic algorithm. We also show that these computational problems are complete for Search-BPP: a pseudodeterministic algorithm for any of these problems implies a pseudodeterministic algorithm for all problems in Search-BPP.

Constraint satisfaction problems (CSP's) and data stream models are two powerful abstractions to capture a wide variety of problems arising in different domains of computer science. Developments in the two communities have mostly occurred independently and with little interaction between them. In this work, we seek to investigate whether bridging the seeming communication gap between the two communities may pave the way to richer fundamental insights. To this end, we focus on two foundational problems: model counting for CSP's and computation of zeroth frequency moments F0 for data streams.

Influence diffusion has been central to the study of the propagation of information in social networks, where influence is typically modeled as a binary property of entities: influenced or not influenced. We introduce the notion of attitude, which, as described in social psychology, is the degree by which an entity is influenced by the information. We present an information diffusion model that quantifies the degree of influence, i.e., attitude of individuals, in a social network. With this model, we formulate and study the attitude maximization problem. We prove that the function for computing attitude is monotonic and sub-modular, and the attitude maximization problem is NP-Hard. We present a greedy algorithm for maximization with an approximation guarantee of (1 - 1/e). Using the same model, we also introduce the notion of “actionable” attitude with the aim to study the scenarios where attaining individuals with high attitude is objectively more important than maximizing the attitude of the entire network. We show that the function for computing actionable attitude, unlike that for computing attitude, is non-submodular but is approximately submodular. We present an approximation algorithm for maximizing actionable attitude in a network. We experimentally evaluated our algorithms and studied empirical properties of the attitude of nodes in the network such as spatial and value distribution of high attitude nodes.

Some of the well known streaming algorithms to estimate number of triangles in a graph stream work as follows: Sample a single triangle with high enough probability and repeat this basic step to obtain a global triangle count. For example, the algorithm due to Buriol et al. (PODS 2006) uniformly at random picks a single vertex v and a single edge e and checks whether the two cross edges that connect$v$to$e$appear in the stream. Similarly, the neighborhood sampling algorithm (PVLDB 2013) attempts to sample a triangle by randomly choosing a single vertex v, a single neighbor$u$of$v$and waits for a third edge that completes the triangle. In both the algorithms, the basic sampling step is repeated multiple times to obtain an estimate for the global triangle count in the input graph stream. In this work, we propose a multi-sampling variant of these algorithms: In case of Buriol et al's algorithm, instead of randomly choosing a single vertex and edge, randomly sample multiple vertices and multiple edges and collect cross edges that connect sampled vertices to the sampled edges. In case of neighborhood sampling algorithm, randomly pick multiple edges and pick multiple neighbors of them. We provide a theoretical analysis of these algorithms and prove that these new algorithms improve upon the known space and accuracy bounds. We experimentally show that these algorithms outperform well known triangle counting streaming algorithms.

We formulate and study Constrained Influence Maximization problem where a network has two types of nodes-targets and non-targets. Given k and θ, the objective is to find a k-size seed set which maximizes the influence spread among the target nodes and keeps the number of non-targets influenced below the threshold θ. The problem, in general, is NP-hard. We also prove that obtaining a constant factor approximation algorithm for this problem is quasi-NP hard. Nevertheless, we are able to present a greedy algorithm and prove that it has certain approximation guarantees with a multiplicative factor of (1 - 1/e) and an additive error, where the latter is dependent on the underlying network structure. We evaluate the extent of the additive error on several representative social networks of varying sizes, and show that in most scenarios, the greedy algorithm indeed provides a high quality solution efficiently. We also develop a multi-greedy algorithm that attempts to keep multiple seed sets and improves upon the greedy algorithm. However, naive implementations of this algorithm is not practically viable due to prohibitively high time overhead. To address this issue, we develop a two-phase heuristic framework to improve the run times. We have conducted extensive empirical evaluation, which not only validates our algorithms, evaluates their effectiveness and efficiency, but also provides important insights on the interplay between the seed-set size, number of non-targets, the threshold, and the additive approximation error on influence-spread.

We investigate the notion of pseudodeterminstic approximation algorithms. A randomized approximation algorithm A for a function f is pseudodeterministic if for every input x there is a unique value v so that A(x) outputs v with high probability, and v is a good approximation of f(x). We show that designing a pseudodeterministic version of Stockmeyer's well known approximation algorithm for the NP-membership counting problem will yield a new circuit lower bound: if such an approximation algorithm exists, then for every k, there is a language in the complexity class ZPP^{NP}_{tt} that does not have n^k-size circuits. While we do not know how to design such an algorithm for the NP-membership counting problem, we show a general result that any randomized approximation algorithm for a counting problem can be transformed to an approximation algorithm that has a constant number of influential random bits. That is, for most settings of these influential bits, the approximation algorithm will be pseudodeterministic.

We formalize notions of triangle and open-wedge heavy-hitters in large networks. Intuitively, a node of a network G is a triangle heavy-hitter if it participates in relatively many triangles of G (analogously for open wedges). These notions have applications in social network analysis. We consider the triangle and open-wedge heavy-hitter problems: the computational problems of maintaining a set of nodes of a network that participate in many triangles and open wedges. We give sampling-based algorithms for these problems when the input network G comes as an edge stream. We prove theoretical guarantees on the quality of solutions, time and space complexity of these algorithms. This is the first work that studies the triangle and open-wedge heavy hitters problem on massive streaming networks. We evaluate the performance of our proposed algorithms by running on several real-world data sets. These experiments indicate that our algorithms efficiently detect heavy hitters while keeping both the false-positive and the false-negative errors very low.

Investigating the complexity of randomized space-bounded machines that are allowed to make multiple passes over the random tape has been of recent interest. In particular, it has been shown that derandomizing such probabilistic machines yields a weak but new derandomization of probabilistic time-bounded classes. In this paper we further explore the complexity of such machines. In particular, as our main result we show that for any epsilon<1, every language that is accepted by an O(n^epsilon)-pass, randomized logspace machine can be simulated in deterministic logspace with linear amount of advice. This result extends an earlier result of Fortnow and Klivans who showed that RL is in deterministic logspace with linear advice.

Automated testing has been a focus of research for a long time. As such, we tend to think about this in a coverage centric manner. Testing budgets have also driven research such as prioritization and test selection, but as a secondary concern. As our systems get larger, are more dynamic, and impact more people with each change, we argue that we should switch from a coverage centric view to a budgeted testing centric view. Researchers in other fields have designed approximation algorithms for such budgeted scenarios and these are often simple to implement and run. In this paper we present an exemplar study on combinatorial interaction testing (CIT) to show that a budgeted greedy algorithm, when adapted to our problem for various budgets, does almost as well coverage-wise as a state of the art greedy CIT algorithm, better in some cases than a state of the art simulated annealing, and always improves over random. This suggests that we might benefit from switching our focus in large systems, from coverage to budgets.

We study the Minimum Circuit Size Problem (MCSP): given the truth-table of a Boolean function f and a number k, does there exist a Boolean circuit of size at most k computing f? This is a fundamental NP problem that is not known to be NP-complete. Previous work has studied consequences of the NP-completeness of MCSP. We extend this work and consider whether MCSP may be complete for NP under more powerful reductions. We also show that NP-completeness of MCSP allows for amplification of circuit complexity. We show the following results. - If MCSP is NP-complete via many-one reductions, the following circuit complexity amplification result holds: If NP cap co-NP requires 2^n^{Omega(1)-size circuits, then E^NP requires 2^Omega(n)-size circuits. - If MCSP is NP-complete under truth-table reductions, then EXP neq NP cap SIZE(2^n^epsilon) for some epsilon> 0 and EXP neq ZPP. This result extends to polylog Turing reductions.

We obtain the following new simultaneous time-space upper bounds for the directed reachability problem. (1) A polynomial-time, O(n^{2/3} * g^{1/3})-space algorithm for directed graphs embedded on orientable surfaces of genus g. (2) A polynomial-time, O(n^{2/3})-space algorithm for all H-minor-free graphs given the tree decomposition, and (3) for K_{3,3}-free and K_5-free graphs, a polynomial-time, O(n^{1/2 + epsilon})-space algorithm, for every epsilon > 0. For the general directed reachability problem, the best known simultaneous time-space upper bound is the BBRS bound, due to Barnes, Buss, Ruzzo, and Schieber, which achieves a space bound of O(n/2^{k * sqrt(log(n))}) with polynomial running time, for any constant k. It is a significant open question to improve this bound for reachability over general directed graphs. Our algorithms beat the BBRS bound for graphs embedded on surfaces of genus n/2^{omega(sqrt(log(n))}, and for all H-minor-free graphs. This significantly broadens the class of directed graphs for which the BBRS bound can be improved.

We show that there is a language that is Turing complete for NP but not many-one complete for NP, under a worst-case hardness hypothesis. Our hypothesis asserts the existence of a non-deterministic, double-exponential time machine that runs in time O(2^2^n^c) (for some c > 1) accepting Sigma^* whose accepting computations cannot be computed by bounded-error, probabilistic machines running in time O(2^2^{beta * 2^n^c) (for some beta > 0). This is the first result that separates completeness notions for NP under a worst-case hardness hypothesis.

The number of triangles in a graph is a fundamental metric widely used in social network analysis, link classification and recommendation, and more. In these applications, modern graphs of interest tend to both large and dynamic. This paper presents the design and implementation of a fast parallel algorithm for estimating the number of triangles in a massive undirected graph whose edges arrive as a stream. Our algorithm is designed for shared-memory multicore machines and can make efficient use of parallelism and the memory hierarchy. We provide theoretical guarantees on performance and accuracy, and our experiments on real-world datasets show accurate results and substantial speedups compared to an optimized sequential implementation.

We show that the reach ability problem over {\em directed planar graphs} can be solved simultaneously in polynomial time and approximately $O(\sqrt{n})$ space. In contrast, the best space bound known for the reach ability problem on general directed graphs with polynomial running time is $O(n/2^{\sqrt{\log n}})$.

This paper presents a new space-efficient algorithm for counting and sampling triangles--and more generally, constant-sized cliques--in a massive graph whose edges arrive as a stream. Compared to prior work, our algorithm yields significant improvements in the space and time complexity for these fundamental problems. Our algorithm is simple to implement and has very good practical performance on large graphs.

In many stream monitoring situations, the data arrival rate is so high that it is not even possible to observe each element of the stream. The most common solution is to sample a small fraction of the data stream and use the sample to infer properties and estimate aggregates of the original stream. However, the quantities that need to be computed on the sampled stream are often different from the original quantities of interest and their estimation requires new algorithms. We present upper and lower bounds (often matching) for estimating frequency moments, support size, entropy, and heavy hitters of the original stream from the data observed in the sampled stream.

This paper presents the following results on sets that are complete for $\NP$. \begin{enumerate} \item If there is a problem in $\NP$ that requires $\twonO$ time at almost all lengths, then every many-one NP-complete set is complete under length-increasing reductions that are computed by polynomial-size circuits. \item If there is a problem in $\CoNP$ that cannot be solved by polynomial-size nondeterministic circuits, then every many-one complete set is complete under length-increasing reductions that are computed by polynomial-size circuits. \item If there exist a one-way permutation that is secure against subexponential-size circuits and there is a hard tally language in $\NP \cap \CoNP$, then there is a Turing complete language for $\NP$ that is not many-one complete. \end{enumerate} Our first two results use worst-case hardness hypotheses whereas earlier work that showed similar results relied on average-case or almost-everywhere hardness assumptions. The use of average-case and worst-case hypotheses in the last result is unique as previous results obtaining the same consequence relied on almost-everywhere hardness results.

We clarify the role of Kolmogorov complexity in the area of randomness extraction. We show that a computable function is an almost randomness extractor if and only if it is a Kolmogorov complexity extractor, thus establishing a fundamental equivalence between two forms of extraction studied in the literature: Kolmogorov extraction and randomness extraction. We present a distribution ${\cal M}_k$ based on Kolmogorov complexity that is complete for randomness extraction in the sense that a computable function is an almost randomness extractor if and only if it extracts randomness from ${\cal M}_k$.

We show that a set is m-autoreducible if and only if it is m-mitotic. This solves a long-standing open question in a surprising way. As a consequence of this unconditional result and recent work by Glaßer et al., complete sets for all of the following complexity classes are m-mitotic: $\mathrm{NP}$, $\mathrm{coNP}$, $\oplus\mathrm{P}$, $\mathrm{PSPACE}$, and $\mathrm{NEXP}$, as well as all levels of $\mathrm{PH}$, $\mathrm{MODPH}$, and the Boolean hierarchy over $\mathrm{NP}$. In the cases of $\mathrm{NP}$, $\mathrm{PSPACE}$, $\mathrm{NEXP}$, and $\mathrm{PH}$, this at once answers several well-studied open questions. These results tell us that complete sets share a redundancy that was not known before. In particular, every $\mathrm{NP}$-complete set A splits into two $\mathrm{NP}$-complete sets $A_1$ and $A_2$. We disprove the equivalence between autoreducibility and mitoticity for all polynomial-time-bounded reducibilities between 3-tt-reducibility and Turing-reducibility: There exists a sparse set in $\mathrm{EXP}$ that is polynomial-time 3-tt-autoreducible, but not weakly polynomial-time T-mitotic. In particular, polynomial-time T-autoreducibility does not imply polynomial-time weak T-mitoticity, which solves an open question by Buhrman and Torenvliet.

Efficient one‐pass estimation of $F_0$, the number of distinct elements in a data stream, is a fundamental problem arising in various contexts in databases and networking. We consider range‐efficient estimation of $F_0$: estimation of the number of distinct elements in a data stream where each element of the stream is not just a single integer but an interval of integers. We present a randomized algorithm which yields an (ε, δ)‐approximation of $F_0$, with the following time and space complexities (n is the size of the universe of the items): (1) The amortized processing time per interval is $O(\log{\frac{1}{\delta}}\log \frac{n}{\epsilon})$. (2) The workspace used is $O(\frac{1}{\epsilon^2}\log{\frac{1}{\delta}}\log n)$ bits. Our algorithm improves upon a previous algorithm by Bar‐Yossef, Kumar and Sivakumar [Proceedings of the $13$th ACM–SIAM Symposium on Discrete Algorithms (SODA), 2002, pp. 623–632], which requires $O(\frac{1}{\epsilon^5} \log{\frac{1}{\delta}}\log^5 n)$ processing time per item. This algorithm can also be used to compute the max‐dominance norm of a stream of multiple signals and significantly improves upon the previous best time and space bounds by Cormode and Muthukrishnan [Proceedings of the $11$th European Symposium on Algorithms (ESA), Lecture Notes in Comput. Sci. 2938, Springer, Berlin, 2003, pp. 148–160]. This algorithm also provides an efficient solution to the distinct summation problem, which arises during data aggregation in sensor networks [Proceedings of the 2nd International Conference on Embedded Networked Sensor Systems, ACM Press, New York, 2004, pp. 250–262, Proceedings of the $20$th International Conference on Data Engineering (ICDE), 2004, pp. 449–460].

We study several properties of sets that are complete for NP. We prove that if L is an NP‐complete set and S \not\supseteq L is a p‐selective sparse set, then $L - S$ is $\leq^{p}_{m}$‐hard for NP. We demonstrate the existence of a sparse set $S \in \mathrm{DTIME}(2^{2^{n}})$ such that for every $L \in \mbox{NP} - \mbox{P}$, L - S is not $\leq^p_m$‐hard for NP. Moreover, we prove for every $L \in \mbox{NP} - \mbox{P}$ that there exists a sparse $S \in $ EXP such that L - S is not $\leq^p_m$‐hard for NP. Hence, removing sparse information in P from a complete set leaves the set complete, while removing sparse information in EXP from a complete set may destroy its completeness. Previously, these properties were known only for exponential time complexity classes. We use hypotheses about pseudorandom generators and secure one‐way permutations to derive consequences for longstanding open questions about whether NP‐complete sets are immune. For example, assuming that pseudorandom generators and secure one‐way permutations exist, it follows easily that NP‐complete sets are not p‐immune. Assuming only that secure one‐way permutations exist, we prove that no NP‐complete set is DTIME$(2^{n^{\epsilon}})$‐immune. Also, using these hypotheses we show that no NP‐complete set is quasi‐polynomial‐close to P. We introduce a strong but reasonable hypothesis and infer from it that disjoint Turing‐complete sets for NP are not closed under union. Our hypothesis asserts the existence of a UP‐machine M that accepts $0^*$ such that for some $0 < \epsilon < 1$, no $2^{n^{\epsilon}}$ time‐bounded machine can correctly compute infinitely many accepting computations of M. We show that if $\UP \cap \co\UP$ contains DTIME$(2^{n^{\epsilon}})$‐bi‐immune sets, then this hypothesis is true.

Efficient one-pass computation of F/sub 0/, the number of distinct elements in a data stream, is a fundamental problem arising in various contexts in databases and networking. We consider the problem of efficiently estimating F/sub 0/ of a data stream where each element of the stream is an interval of integers. We present a randomized algorithm which gives an (/spl epsiv/, /spl delta/) approximation of F/sub 0/, with the following time complexity (n is the size of the universe of the items): (1) the amortized processing time per interval is O(log1//spl delta/ log n//spl epsiv/). (2) The time to answer a query for F/sub 0/ is O(log1//spl delta/). The workspace used is O(1//spl epsiv//sup 2/log1//spl delta/logn) bits. Our algorithm improves upon a previous algorithm by Bar-Yossef Kumar and Sivakumar (2002), which requires O(1//spl epsiv//sup 5/log1//spl delta/log/sup 5/n) processing time per item. Our algorithm can be used to compute the max-dominance norm of a stream of multiple signals, and significantly improves upon the current best bounds due to Cormode and Muthukrishnan (2003). This also provides efficient and novel solutions for data aggregation problems in sensor networks studied by Nath and Gibbons (2004) and Considine et. al. (2004).

We study several properties of sets that are complete for NP. We prove that if L is an NP-complete set and S /spl nsupe/ L is a p-selective sparse set, then L -S is /spl les//sub m//sup p/-hard for NP. We demonstrate existence of a sparse set S /spl isin/ DTIME(2/sup 2n/) such that for every L /spl isin/ NP - P, L - S is not /spl les//sub m//sup p/-hard for NP. Moreover, we prove for every L /spl isin/ NP - P, that there exists a sparse S /spl isin/ EXP such that L - S is not /spl les//sub m//sup p/-hard for NP. Hence, removing sparse information in P from a complete set leaves the set complete, while removing sparse information in EXP from a complete set may destroy its completeness. Previously, these properties were known only for exponential time complexity classes. We use hypotheses about pseudorandom generators and secure one-way permutations to derive consequences for long-standing open questions about whether NP-complete sets are immune. For example, assuming that pseudorandom generators and secure one-way permutations exist, it follows easily that NP-complete sets are not p-immune. Assuming only that secure one-way permutations exist, we prove that no NP-complete set is DTIME(2/sup ne/)-immune. Also, using these hypotheses we show that no NP-complete set is quasipolynomial-close to P. We introduce a strong but reasonable hypothesis and infer from it that disjoint Turing-complete sets for NP are not closed under union. Our hypothesis asserts existence of a UP-machine M that accepts 0* such that for some 0 < /spl epsi/ < 1, no 2/sup ne/ time-bounded machine can correctly compute infinitely many accepting computations of M, We show that if UP /spl cap/ coUP contains DTIME(2/sup ne/)-bi-immune sets, then this hypothesis is true.

The Turing and many-one completeness notions for NP have been previously separated under measure, genericity, and bi-immunity hypotheses on NP. The proofs of all these results rely on the existence of a language in NP with almost everywhere hardness. In this paper we separate the same NP-completeness notions under a partial bi-immunity hypothesis that is weaker and only yields a language in NP that is hard to solve on most strings. This improves the results of Lutz and Mayordomo (1996), Ambos-Spies and Bentzien (2000), and Pavan and Selman (2002). The proof of this result is a significant departure from previous work.

We show that if SAT does not have small circuits, then there must exist a small number of formulas such that every small circuit fails to compute satisfiability correctly on at least one of these formulas. We use this result to show that if P/sup NP[1]/=P/sup NP[2]/, then the polynomial-time hierarchy collapses to S/sub 2//sup P//spl sube//spl Sigma//sub 2//sup p//spl cap//spl Pi//sub 2//sup p/. Even showing that the hierarchy collapsed to /spl Sigma//sub 2//sup p/ remained open.

We use hypotheses of structural complexity theory to separate various NP-completeness notions. In particular, we introduce a hypothesis from which we describe a set in NP that is /spl les//sub T//sup P/-complete but not /spl les//sub tt//sup P/-complete. We provide fairly thorough analyses of the hypotheses that we introduce.

We use hypotheses of structural complexity theory to separate various NP-completeness notions. In particular, we introduce an hypothesis from which we describe a set in NP that is $\mbox{${\leq}^{\rm P}_{\rm T}$}$-complete but not $\mbox{${\leq}^{\rm P}_{tt}$}$-complete. We provide fairly thorough analyses of the hypotheses that we introduce.